Description

Intro

Telesat (TSAT) was last written up in December 2021 and much has changed since then. At a high level, the geostationary (GEO) satellite business has continued to deteriorate with likely no residual equity value remaining and only partial recovery for bondholders. Meanwhile, the low earth orbit (LEO) business continues to advance towards commercialization with several milestones having been achieved over the past 12 months including securing terms for the Government of Canada financing and contracting with MDA as the prime manufacturer at attractive rates. The company does not screen well on consolidated financial metrics and requires a detailed understanding of the balance sheet to appreciate the substantial value imbedded in the stock. Assuming the GEO business is worthless, the unrestricted LEO subsidiary is currently being valued at C$503 mm (TSAT’s market cap) with no debt and a cash position of C$1.2 Bn. This also assigns no value for the C$540 mm spent on PP&E thus far along with the valuable IP and regulatory approvals held within LEO. We believe TSAT is particularly actionable given near term catalysts we anticipate in the upcoming Q1/24 release (see below).

Highlights                                                                                                                                                                                                                                                                                               
- When looking at TSAT’s consolidated financial results, it is readily apparent why the company garners limited interest from public investors. The legacy geostationary satellite (GEO) business which currently generates all of the company’s cash flow is faced with an overwhelming debt burden and rapid secular decline in direct-to-home television. TSAT also has limited analyst coverage and a very limited public float with PSP Investments and MHR collectively owning 72% of the company. What is lost in TSAT’s consolidated financials is its highly promising Lightspeed low earth orbit (LEO) subsidiary which continues to move toward commercialization. TSAT deliberately structured its balance sheet to retain full ownership of the LEO business while immunizing shareholders from the distressed GEO subsidiary where the large debt burden resides.

- Competing against heavyweights like Amazon and SpaceX, it is easy to discount the prospects of TSAT’s Lightspeed LEO project. The fact that Starlink is already capacity constrained with a subscriber base of only ~2.6 mm is indicative of the robust demand for LEO services and highlights the need for substantial capacity growth from all sources. In addition, LEO broadband is unlikely to be a winner-take-all market given important national security and antitrust considerations which should support the success of smaller players like TSAT and OneWeb. Lastly, according to third party research from MIT’s Aeronautics and Astronautics Department which reviewed the FCC filings of the four key LEO projects, TSAT Lightspeed is highly competitive with total system bandwidth of 5.0 to 7.5 Tbps (depending how much you discount it to reflect the 2022 design change) expected to be up to 5x greater than OneWeb and only 26% less than that of Starlink. On a revenue per capacity basis, this research suggests TSAT’s ambitious long term LEO revenue guidance is achievable.

- While there may be no residual equity value in the GEO business, we see significant value in TSAT. At current levels, the market is valuing the LEO subsidiary at a 55% discount to the unrestricted cash held on its balance sheet while assigning no value to the C$540 mm invested in PP&E along with valuable IP and regulatory approvals. Based on a simple DCF model which is more conservative than management guidance, we estimate fair market value of the LEO business of ~US$2.5bn or US$49 per TSAT share (15% WACC). Of note, this number is not too dissimilar from the US$3 Bn strike price set for the Government of Canada (GoC) warrants.

- We see two near term catalysts for TSAT. First, the company announced terms for its GoC financing on April 1st. Incremental colour in the Q1/24 release on closing this financing should support TSAT equity as this is a major de-risking event. Second, based on publicly available trading volume for TSAT’s bonds, we believe the company has repurchased a substantial amount of GEO debt at an attractive price in April potentially in the ~US$150 mm range. Importantly, the cash used to purchase this debt was within the GEO subsidiary as the company is unable to move cash from LEO to GEO. While this does not alter our assumption that the GEO business has no residual equity value, rapidly reducing GEO debt nevertheless cleans up the consolidated financials and allows for greater focus on TSAT Lightspeed.

- Deutsche Bank launched coverage on TSAT equity on April 29th with a Hold rating and US$9 TP. We think this is generally a positive development as any incremental exposure is positive. We disagree with two major assumptions in the DB initiation. First, in the valuation the analyst implicitly assumes the equity value of GEO is negative $1.7 Bn by assuming the company will voluntarily repay all of the GEO debt (at face value no less). Second, they apply a 10.2x terminal multiple to 2030 EBITDA of US$694 mm which is 52% below management guidance and 74% below the peak run-rate EBITDA expected in 2032. While some conservatism is clearly warranted, this seems excessive in our view. 

GEO Business Overview                                                                                                                                                                                                                                                                       
While the long-term decline of TSAT’s GEO business is well understood, the pace at which the business has deteriorated in recent quarters has been a surprise both to management and the market. The GEO business faces a combination of secular decline in direct-to-home television satellite service and disruption in the enterprise business from LEO competitors Starlink and OneWeb. At this point, the company is rapidly depleting its contracted GEO backlog which amounted to C$1.3 Bn at the end of 2023. Of critical importance, TSAT’s carefully designed corporate structure immunizes TSAT equity from the GEO subsidiary where the company’s substantial C$3.1 Bn distressed debt stack resides. Under the likely scenario that there is no residual equity value remaining within the GEO business, the fundamental value of TSAT shares is derived entirely from its unrestricted LEO subsidiary. That being said, the GEO business continues to have an outsized impact on TSAT equity sentiment as its large debt burden and declining profitability are reflected in the consolidated financial results. For this reason, we believe the extent to which the company is able to repurchase debt at a significant discount to par within the GEO silo is positive as it improves consolidated metrics and simplifies the story. Below we highlight the company’s GEO backlog, restricted cash position and gross debt over the past two years which reflect this dynamic. 

TSAT GEO Backlog, Gross Debt & Restricted Cash

Over the past two years, TSAT has repurchased US$587 mm of debt for proceeds of US$332 mm implying an average 43% discount to face value. These repurchases have been evenly spread across the four outstanding debt tranches. The rate of debt repurchases slowed for the past three quarters, causing some concern for debt holders. Management noted that this pause was not a reflection of changing priorities but rather was in response to creeping takeover provisions that prevented the company from repurchasing more than 25% of its debt in the open market. In addition, the company felt it could not repurchase debt until it finalized and press released terms of the Federal Government funding package for LEO as announced on April 1, 2024. Since this date, trading volume for the company’s notes has picked up significantly (US$ 256 mm volume in the first three weeks of April vs US$120 mm in all of Q1/24) corroborating our view that the company is back in the market repurchasing bonds which should be confirmed in the Q1/24 earnings release in May. Inclusive of Term Loan B repurchases, we would not be surprised to see ~US$150 mm of debt reduction reported for April.

In terms of the GEO business itself, the company has been consuming its contracted backlog with spot sales accounting for just 20% of revenue over the past three years. At the end of 2023, TSAT’s remaining GEO backlog stood at C$1.3 Bn which was down 38% or C$0.8 Bn from two years prior. While the company has secured some new bookings to the backlog over this period, at the current pace of decline the backlog is expected to be largely exhausted over the next three years. Even assuming the company is able to maintain its high FCF conversion of ~80% of revenue, it appears unlikely that there is any material equity value remaining within the GEO business.

TSAT Debt Repurchases (2022 to Present)

                                              Source: Company filings

LEO Competitive Landscape: Four Player Market Unfolding                                                                                                                                                                                                        
Given the long and complex U.S. Federal Communications Commission (FCC) regulatory process and significant upfront capital required to launch LEO satellite services, we have relatively clear visibility as to how the LEO competitive landscape will unfold in the coming years. LEO operators are assigned specific orbits and radio frequencies within which they can operate their constellations to prevent overcrowding and interference. As a result, there is incumbent value to owning real estate in orbital space as it creates barriers of entry for new competitors. There are only two players active in the LEO broadband market today: Starlink and OneWeb. Despite its dominant market position, Starlink itself is a relatively new having only launched its service in 2021. OneWeb is even newer, having launched most of its satellites in the 2021-23 timeframe and the company is still completing the buildout of its surface infrastructure for full commercialization. Iridium Communications is technically a third player in the LEO broadband market via its NEXT project which launched in 2017-18. However, this project is designed for basic marine and aviation voice communication capabilities with max bandwidth of 704 Kbps and is thus not a true competitor to Starlink and OneWeb.

Low Earth Orbit In-Service Dates

       Source: Company filings, Media Reports

Among the future projects in the queue, there are only two additional competitors expected to enter the market over the next four years being Kuiper (Amazon) and Lightspeed (TSAT) due in ~2027 and ~2028 respectively. While Kuiper and Starlink clearly have substantial competitive advantages given their vast financial resources and close ties to space transport providers (Blue Origin and SpaceX), this is unlikely to be a winner-take-all market given the important national security and antitrust issues at play which should support the success of OneWeb and Lightspeed particularly in the enterprise and government sectors. Lastly, it’s worth noting that China is working on a large state-owned project named Hongyan which is rumoured to be targeting an in service date of ~2029. However, as a Chinese state owned entity, geopolitics and privacy concerns suggest it is likely not a direct competitor to TSAT.

In mid-2021 the MIT Aeronautics and Astronautics Department released an updated version of its research paper comparing the current and future generation offerings from SpaceX, OneWeb, TSAT and Amazon based on FCC filings at the time (http://systemarchitect.mit.edu/docs/pachler21a.pdf). Of note, MIT’s system throughput estimates suggested TSAT’s total bandwidth would be >5x greater than that of OneWeb’s existing constellation and only 27% less than that of Starlink. Subsequent to the release of this report TSAT reduced the size of its constellation by a third and thus we are left guessing where proforma capacity sits within a range of 530 and 7.5 Tbps. These figures will also evolve as the various constellations are upgraded and modernized over time. Nevertheless, this does provide context to the notion that TSAT’s project is highly competitive with peers despite its relatively low satellite count due to the use of much larger and more sophisticated satellites that promote greater system efficiency. This research paper also helps validate TSAT’s revenue guidance for Lightspeed which appears reasonable on a price per capacity basis relative to peers even when discounting MIT’s system throughput estimate by a third.

LEO Constellation Analysis (as of Mid-2021)

                                                             Source: MIT Aeronautics and Astronautics Department

Starlink (SpaceX):
Starlink was the first mover and remains the clear market leader in LEO broadband with 2.6 mm subscribers and >95% market share today. The company began deploying its constellation in 2019 with its service offering officially launching in early 2021. The project is primarily designed to be a direct-to-consumer offering which targets the estimated 400 to 500 million households that are unable to access ground-based fiber given cost, complexity and/or geography. Starlink has clear competitive advantages within the direct-to-consumer market given the strong brand awareness tied to SpaceX founder Elon Musk. Starlink is also strategic for SpaceX as it provides steady and rateable business for the core space transport operation. Starlink’s product offering is very broad with several different tiers that cater to customers unique requirements in terms of speed and mobility. These include Starlink Residential (fixed location), Starlink RV, Starlink Maritime and Starlink Business. In late 2022, the company also launched a specialized service for government and military agencies called Starshield which operates a separate satellite network (albeit heavily integrated with the existing constellation). Starshield was recently awarded the first contract issued from the U.S. Space Force for LEO services.

Starlink Subscriber Count

                                       Source: Company Disclosure

Starlink has encountered two key challenges during its commercialization: bandwidth capacity constraints and supply chain challenges meeting demand for user terminals. On the first point, the company found that the rapid increase in subscribers in certain regions has led to deteriorating internet speed as the company quickly ran out of available bandwidth capacity. The median download speed for U.S. Starlink users at launch was >100 Mbps and quickly fell to the 60 to 80 Mbps range as the subscriber count grew exponentially. In response, the company has had to temporarily cap the number of users in these high demand regions, creating long customer waitlists. As a consolation, Starlink launched a “best efforts” service which allows customers on a waitlist to subscribe to Starlink with a low-priority connection that is throttled back when system capacity is limited. The fact that Starlink is already capacity constrained with a subscriber base of only ~2.6 mm is indicative of the robust demand for LEO services and highlights the need for substantial capacity growth from Starlink and other players. Given these data capacity constraints, Starlink has created a very regional and dynamic pricing framework which sets localized pricing based on the overall demand within a geography relative to the available capacity.

OneWeb (EUTelsat):
OneWeb’s 1st generation LEO project was a victim of both the Covid-19 pandemic and the Russia-Ukraine war which led to significant delays and cost overruns. The company had contracted Russian space transport providers to deploy its satellites resulting in substantial delays and lost equipment following the sanctioning of Russian state enterprises. Ultimately OneWeb went through a bankruptcy process and was acquired by European GEO satellite operator EUTelsat in 2023 as a means to enter the LEO market. Since then, EUTelsat has been focused on bringing the OneWeb constellation to market with all required satellites now in orbit; the remaining item delaying full commercialization is the rollout of the ground station station network which is required to provide global coverage. This is anticipated to occur by the end of 2024. OneWeb’s backlog has nevertheless been rapidly growing, currently at €1.1 Bn (+23% Q/Q).

Importantly, given years of delays, OneWeb’s first generation constellation lags peers in terms of performance specifications with a key limiting factor being the inability of its satellites to connect to each other via optical inter-satellite links (ISL). Without this capability, a higher number of ground stations are required and the overall efficiency of the system is much more limited. Recognizing these shortcomings, management has been working on a second generation constellation which was originally set to launch around 2028 to compete with Kuiper and Lightspeed. In light of the challenging debt capital markets, the company has since revised its strategy and now expects to continue operating its existing constellation until ~2030. Management has justified this change by stating that a complex migration to the second generation constellation during the launch of Kuiper and Lightspeed LEO service may lead to unnecessary customer churn.

Kuiper (Amazon):
Amazon’s LEO project is similar to Starlink in that it is primarily a direct-to-consumer offering. Amazon expects to deploy “greater than US$10 Bn” of capital on Kuiper with the first phase including 578 satellites. The company’s commercialization timeline remains relatively opaque though it has noted that it expects deployment of satellites to commence in 1H/24 with early customer pilots to begin by the end of 2024. Similar to Starlink, it will likely take two to three years of launches before Kuiper reaches commercial scale. The company’s FCC license requires that at least 50% of its satellites are launched by mid-2026. Amazon expects to have three tiered service levels each with its own user terminal. The compact terminal will offer maximum speeds of up to 100 Mbps with the standard terminal at 400 Mbps and the enterprise terminal at 1 Gbps. The company has yet to announce pricing details for these tiers but noted that affordability is a key principle and the cost will vary considerably depending on region and customer type.

Valuation                                                                                                                                                                                                                                                                                                
At first glance, TSAT screens poorly from a valuation perspective given the excessive leverage and declining EBITDA associated with the GEO business which is reflected in the consolidated financials. TSAT is encumbered with C$3.2 Bn of debt comprised of a US$1.42 Bn term loan, US$0.7 Bn of senior secured notes and US$0.3 Bn of unsecured notes. This is partially offset by a large cash position of C$1.7 Bn yielding total net debt of C$1.5 Bn (~4.3x debt/EBITDA). What is lost in TSAT’s consolidated financials is the fact that the company has very deliberately structured its balance sheet in order to ring fence the valuable LEO subsidiary to protect it from the overleveraged and declining GEO business. Of the C$1.7 Bn of cash TSAT holds, C$1.2 Bn is held within the LEO subsidiary. In essence, when buying TSAT equity at current prices, investors are buying the cash-rich LEO shell for C$0.40 on the dollar while assigning no value to the C$540 mm spent on PP&E thus far along with the valuable IP and regulatory approvals held by TSAT LEO.

TSAT LEO Revenue by Category

                     Source: Company Filings

Despite all the progress made to date, TSAT’s Lightspeed LEO project is still early stage with revenue not anticipated until 2028. The company recently released a detailed financial forecast in its November 2023 slide deck which assumes total capital spend of US$3.9 Bn over the next several years (>90% fixed price contracts) with 2024 being the year of peak spend at ~US$1 Bn. The company has included US$400 mm of contingency in its capital program. With a gradual ramp up in sales, management anticipates LEO reaching peak run-rate revenue and EBITDA of US$3.2 Bn and US$2.7 Bn respectively in 2032 reflecting ~1% market share in the enterprise TAM. While its difficult to validate these numbers, there are three supporting factors which provide some incremental comfort with this guidance:

1. Revenue Per Capacity Ratio Appears Conservative: With Starlink and OneWeb active in the market, we have some indication as to the revenue potential of the LEO business. Based on recent media reports, Starlink is expected to generate ~US$7 Bn of revenue this year (and growing) while OneWeb recently guided to stabilized revenue of >€1 Bn. Using third party technical estimates from MIT, each of these figure’s maps to ~US$0.7 Bn of revenue per Tbps of system capacity. As MIT has not updated its estimates since TSAT reduced its proposed satellite count by a third, the 7.52 Tbps system capacity estimate is likely too high. However, the capacity reduction should be proportionately less than the satellite count decrease. In any event, TSAT’s Lightspeed guidance implies stabilized revenue per capacity of between US$0.4 Bn and US$0.6 Bn per Tbps which is 22% less than that of Starlink and OneWeb at the midpoint. Of course the price of data will undoubtedly decline over time and thus some conservatism is warranted given TSAT’s long timeline to commercialization.

2. Sophisticated Stakeholder Involvement: TSAT has several large and sophisticated stakeholders engaging with the company including the Canadian Federal Government, PSP Investments and MHR. The U.S. FCC will also have completed a detailed review of the Lightspeed project including from an economic feasibility standpoint. All of these stakeholders should have intimate knowledge of Lightspeed having completed extensive due diligence with far greater access to information than is available to non-insiders.

3. Secured Backlog: In terms of commerciality of the project, TSAT has already secured US$0.6 Bn in backlog for Lightspeed despite the fact the service won’t be available for another four years. This is particularly impressive in the context of OneWeb who launched their project several years ahead of TSAT and is operational today yet has a backlog of only €1.1 Bn.
   
   

LEO Constellation Comparison (Revenue & System Capacity)

                                             Source: MIT Aeronautics and Astronautics Dept, Company Filings, Media Reports

We have built a basic DCF model underpinned by management’s capex and revenue guidance coupled with our assumptions of maintenance capital costs, cash taxes and a 10-year useful life. We have assumed more conservative peak EBITDA margins of 70% (as opposed to TSAT’s guided 84%) to be more consistent with the mature and more competitive GEO business. Assuming a conservative 15% WACC, this maps to a NPV of the LEO business of US$2.5 Bn or US$49/Sh inclusive of the ~US$900 mm unrestricted cash position. Of note, as part of its Government of Canada financing, TSAT issued warrants for 10% of the shares of TSAT LEO at an equity valuation of US$3 Bn which is not too dissimilar from our figure. While we should not underestimate the significant execution risk associated with the Lightspeed project, TSAT shares appear highly compelling from a risk-reward perspective given the substantial margin of safety imbedded in the valuation at current levels.

TSAT LEO DCF Model

Source: Internal Estimates

Risks                                                                                                                                                                                                                                                                                                       
Execution Risk: TSAT’s Lightspeed project is still relatively early in its development and there is significant execution risk ahead. In the near term, the key risk is closing the Government of Canada financing without which the project may be unviable. The primary outstanding condition required to close this financing is to complete definitive agreements with other financing sources to the Government of Canada’s satisfaction. Beyond this, there is of course significant execution risk both from a technical perspective (successful launch and deployment of the constellation and ground station network) as well as from a cost perspective as projects of this magnitude rarely complete on budget. Given the restricted nature of the GEO cash flows, incremental cost overruns would likely need to be funded through additional equity and/or debt financings which could dilute project economics.

Complex Corporate Structure: Thus far TSAT has been very deliberate about preserving the value of the LEO business and protecting it from headwinds in the GEO business. There are also significant restrictions in place that limit management’s ability to move cash between the restricted and non-restricted subsidiary. Nonetheless, the company has faced relentless pressure and lobbying from bondholders to support the GEO debt load with LEO cash flow in the future. At this point it is unclear if management is ready and willing to allow the GEO business go through a formal restructuring process as there may be strategic reasons to avoid doing so. Since the company’s senior notes are guaranteed by the parent company, the exact mechanism through which TSAT would restructure the GEO subsidiary is unclear. The most likely scenario may be to spinout the LEO business to TSAT shareholders prior to a formal restructuring.

Technological Obsolescence: Technology in the LEO market is rapidly advancing as is evidenced when comparing the technological specifications of Iridium NEXT, OneWeb and Starlink all of which launched within just a five-year window. If the rapid pace of technological advancement continues, future constellations like Lightspeed may have a shorter useful life and worse project economics than currently envisioned. Alternatively, they may face intense pricing pressure over time as more advanced constellations supersede them.


Disclosure                                                                                                                                                                                                                                                                                               
I do not hold a position with the issuer such as employment, directorship, or consultancy.
I and/or others I advise hold a material investment in the issuer's securities.

Catalyst

Financing Closing: Closing the Government of Canada financing is of critical importance to Lightspeed, without which the viability of the project is in question. Based on commentary on the conference call and discussions with management, this appears to be a key point of consternation for investors. On April 1, TSAT announced it had formalized the terms of the GoC financing. The next step will be to complete definitive agreements for all remaining financing sources in order to close the GoC financing.  

LEO Monetization/IPO: TSAT’s complex corporate structure makes it difficult for the market to appreciate the significant fundamental value imbedded in the LEO subsidiary. While not a formal valuation, it is notable that the Government of Canada agreed to a US$3 Bn strike price for its TSAT LEO warrants which is fairly consistent with the DCF value we have calculated in our DCF analysis. As a next step, taking a small piece of the LEO business public would have two distinct benefits: it would simplify the story by providing a highly visible and public mark for the LEO subsidiary and it would allow TSAT to raise cash at seemingly more attractive terms. Based on our discussions with management, they are supportive of the idea of a LEO IPO in the future but believe it is prudent to wait until the project is further de-risked in order to get a better valuation.

Additional LEO Backlog Secured: TSAT has a US$0.6 Bn LEO backlog which is impressive given how early the project is in its development. As the company moves closer to its targeted in-service date of late 2027, we expect it to secure additional large commitments from enterprise customers which will allow it to build a more substantial backlog and de-risk Lightspeed over time. Given the strong visibility and high margins that this backlog affords, the market should reward the company with an improved multiple.